Uber Interview Questions — Page 16

Uber·Software Engineer·Technical Phone Screen

May 2026

A deep-dive project interview at Uber for a software engineering role, essentially 30 minutes of someone poking holes in one thing you've built. No warmup, no softballs.

Walk me through a project you worked on: what was the problem you were solving and how did you define it?
What other approaches did you consider before going with the one you chose, and why did you rule them out?
What were the key technical decisions and tradeoffs in the project?
What metric were you optimizing for and what result did you actually achieve?
What blockers did you hit during the project and how did you get past them?
What would you do differently if you were starting this project over today?
What was your specific contribution versus what the rest of the team did?

“This is where I fumbled a bit at the start.”

View Post

Uber·Product Manager·Onsite - Product Sense / Strategy

May 2026

PM interview at Uber, product design question focused on ridesharing. Just the one question from what I can tell, but it had enough surface area to go pretty deep.

What parameters would you consider when designing UberPool?

“I went straight to the matching algorithm and rider experience, which felt right, but I totally glossed over the driver side for way too long.”

View Post

Uber·Machine Learning Engineer·Technical Phone Screen

May 2026

Uber ML Engineer interview with a graph traversal coding problem. Pretty standard algorithmic round, nothing too wild, but the follow-up about weighted edges was a nice curveball.

Given n cities connected by bidirectional edges and a starting city, return all cities sorted first by shortest distance from the source, then by city index for ties.

“Went with BFS pretty quickly since the edges seemed unweighted at first glance.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Uber coding interview for a Software Engineer role. The main problem was a recursive expression parser, which is already a pain, but they added a twist where you also have to handle malformed input and return an error indicator. Not a relaxing afternoon.

Given a string representing a Lisp-style expression with let, add, and mult operations, evaluate it and return the result. Additionally, handle syntactically invalid inputs (mismatched parentheses, unknown tokens, etc.) and return an appropriate error indicator.

“The base problem is already one of those recursive parsing questions that looks manageable until you actually try to write it.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Uber SWE coding round, one string problem that looks deceptively simple but has a slick O(n) solution most people probably don't know cold.

Given a string, determine whether it can be formed by repeating a smaller substring of itself multiple times.

“I knew there was a clever trick here but blanked on it mid-interview and just went with the brute force: try every possible prefix length that divides the total length, check if tiling it reproduces the string.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Uber SWE interview, coding round, one meaty grid simulation problem that took up most of the time. Not super brutal but there were enough layers to trip you up if you weren't careful.

A robot starts at (0,0) on an infinite integer grid and receives a sequence of movement commands (U, D, L, R). Some cells are blocked. If a move would land on a blocked cell, the robot stays put. Return the maximum squared Euclidean distance from the origin at any point during the simulation. Handle up to 1e5 commands and 1e5 obstacles efficiently.

“I went straight for the hash set on obstacles, which was the right call, O(1) lookup per move.”

View Post

Uber·Data Scientist·Technical Phone Screen

May 2026

Uber DS interview that was basically a full A/B testing case study crammed into one session. The question was dense, covering everything from hypothesis setup to alpha spending to CUPED variance reduction. Felt more like a take-home exam than a conversation.

For a primary metric defined as trip completion rate per request, state precise null and alternative hypotheses. Should the test be one-sided or two-sided, and why?
Using the provided request and completion counts for both control and treatment groups, compute the z-test statistic, p-value, and a 95% confidence interval for the lift in completion rate. Show your work.
For the wait time guardrail metric, which test should you use and why? Compute the 95% CI for the mean difference and state your distributional assumptions.
Apply Holm-Bonferroni correction across all three outcomes (primary metric, cancel rate guardrail, wait time guardrail) at a familywise alpha of 0.05. Which effects survive?
In plain language, what does the p-value you computed actually mean, and what does it not mean?
The experiment had five unplanned interim looks at the data with no pre-registered alpha spending. How does this affect your significance conclusion, and how would you apply a Pocock or O'Brien-Fleming approach to address it?
If the pre-experiment completion rate has a correlation of 0.40 with the in-experiment outcome, estimate the variance reduction from CUPED and explain how it would affect required sample size or interpretation of results.
Given all the analysis above, including the guardrail results, should you ship the feature, iterate, or stop? Defend your recommendation.

“I went two-sided initially out of habit, then second-guessed myself mid-answer.”

View Post

Uber·Software Engineer·Technical Phone Screen

May 2026

Uber SWE interview with a classic grid search problem. Pretty standard coding round, nothing too surprising, but the backtracking edge cases will get you if you're not careful.

Given a 2D grid of characters and a target word, determine whether the word can be found in the grid by moving through horizontally or vertically adjacent cells, without reusing any cell.

“Classic backtracking problem and I knew it the second I read it.”

View Post

Uber·Software Engineer·Onsite - Behavioral / Leadership

May 2026

Behavioral onsite for a software engineer role at Uber. Five questions, all pretty standard stuff, but a couple of them pushed for real depth on past projects and team dynamics.

Tell me about the most difficult project you've worked on. What made it hard, what was your role, and what was the outcome?
Describe a time you had a conflict with a coworker and how you resolved it.
Why are you looking to leave your current role?
Why Uber specifically? What draws you to this company or team?
Walk us through one of your favorite projects in depth, including design decisions, your contribution, and what you'd change looking back.

“This one took up a lot of the conversation.”

View Post

Uber·Data Scientist·Technical Phone Screen

May 2026

This was a deep-dive case interview for a Data Scientist role at Uber, basically one big question about a past project that you have to dissect from every angle. It's less of a conversation and more of a structured interrogation on whether you actually understand the business impact of your own work.

Pick a project you've worked on and walk through its full business impact: the problem, what decision your analysis enabled, the key metrics, and a quantified counterfactual including dollar impact with baseline, lift, confidence, and sample sizes.
What was your measurement strategy for that project? Was it a true experiment or a quasi-experiment, what assumptions did you make, and how did you validate them?
How did you handle stakeholder pushback or misalignment during the project, and how did you get people to actually adopt your findings?
What trade-offs did you make during the project, what actually failed, and if you ran it again from scratch, what would you do differently?
How did you manage risks and guardrails throughout the project timeline?

“This is where I underestimated how deep they wanted to go.”

View Post